{ "architectures": [ "SiglipVisionModel" ], "initializer_factor": 1.0, "model_type": "siglip_vision_model", "torch_dtype": "float32", "transformers_version": "4.37.1", "hidden_size": 768, "image_size": 384, "intermediate_size": 3072, "num_attention_heads": 12, "num_hidden_layers": 12, "patch_size": 16 }